Cyber-Physical Mobility System: AI Network

Project Overview

This research focuses on developing advanced AI-driven systems for modeling and predicting traffic patterns around work zones and incidents. By leveraging data science techniques for traffic dataset fusion, we aim to enhance safety and traffic flow efficiency in these critical areas.

AI networks offer real-time insights into system conditions, even in areas with limited data due to signal loss or low sensor coverage. They predict traffic by considering factors such as weather and special events, accurately forecasting future traffic states and understanding underlying causes and chain effects. For instance, an AI network can predict how a highway incident might lead to congestion on nearby roads or how severe weather might impact public transit ridership and road traffic.

AI networks form the backbone of our digital mobility infrastructure, powered by deep learning and generative models. These models enable the networks to dynamically evolve and autonomously generate adaptations, like optimized pedestrian paths and bicycle routes based on real-time movement patterns. It also features auto-calibration, continuously fine-tuning its parameters based on observed discrepancies without human intervention.

Moreover, the network's evolution capability enables it to adapt to long-term changes in urban forms, policy, and human behavior, adjusting its model over time to accommodate new infrastructure, shifts in population density, or evolving transportation policies.